Words with AttitudePaper’s GoalOsgood’s Semantic Differential TechniqueUsing WordNet with Osgood’s theoryMPL ExamplesMPLTRIEVA resultsOther scalesEVA*, POT*, ACT*ApplicationAccuracyAccuracy problemsAuthor’s closing notesWords with AttitudeJaap KampsMaarten MarxPaper’s GoalJudge the emotive or affective meaning of a textUse WordNet to determine values of words with Osgood’s semantic differential techniqueOsgood’s Semantic Differential TechniqueJudge words, phrases, texts by asking subjects to rate them on scales of bipolar adjectivesA subject might be asked to rate “proper” on scales like optimistic-pessimistic, serious-humorous, and active-passive.It turns out that good-bad, strong-weak, and active-passive values account for most variance in judgmentUsing WordNet with Osgood’s theoryAuthors want to get values for words from WordNetThey define MPL(w1,w2) as the minimal path length between w1 and w2, using only same-synset relationsAllowing more than just same-synset damages metricMPL ExamplesMPL(good, proper) = 2(good,right,proper)MPL(good, neat) = 3MPL(good, noble) = 4Can we use this to rate “proper”, “neat”, and “noble” on a good-bad scale?MPLMPL(good, bad) = 4If we just look at MPLs, “noble” is as good as “bad”We need to do something a bit more complicatedTRITo determine the good-bad (“evaluative”) value of wi, examine TRI(wi;good,bad)Define EVA(w) = TRI(w;good,bad)),(),(),();(,jkjikikjiwwMPLwwMPLwwMPLwwwTRIEVA resultsThere are 5410 adjectives linked to “good” or “bad”.Average value of EVA for these 5410 words is –0.00891440)(1404)(25.0445)(0433)(1426),(),(),(),;()(ba dEVAgoodEVAno bleEVAneatEVAbadgo odMPLgoodproperMPLbadproperMPLba dgoodpro perTRIproperEVAOther scalesDefine POT as TRI(w;strong,weak)Define ACT as TRI(w;active,passive)EVA, POT, ACT are well-defined for exactly the same set of 5410 adjectives.EVA*, POT*, ACT*Define EVA*(w) to be EVA(w) if a path exists between w and “good”, and 0 if it doesn’tThis gives us a well-defined function for all wDo the same thing to get POT* and ACT*ApplicationWe can now take the sum of EVA*, POT*, ACT* for all words in a text to get an idea of the good-bad, strong-weak, active-passive values for the text as a wholeAccuracyNo corpus existed that had already been rated for these values, so accuracy could not be tested on a large scaleTests on small numbers of Internet discussions show correspondence between results of this method and actual value of texts, but questionable accuracy for short textsWorks better for long textsAccuracy problemsWith longer texts, false positives and false negatives cancel each other out; doesn’t help for shorter textsLonger texts yield scores of higher magnitude, in general – need to normalize scoresApparent bias to positive words (positive opinions more extensively elaborated, affecting a text’s score more than negative opinions)Author’s closing notesAuthors of texts on Internet discussion sites must be less subtle about good/badLittle NLP research addresses subjective aspects; this paperhelps fill the
View Full Document